Back

Targeted Interventions to Improve Discharge Times in an Australian Cardiology Department: a Before-and-After Study.

Xiong, E.; Da Silva, D.; Elsing, M.; Perry, D.; Elridge, M.; Carberry, A.; Decourcy, V.; King, Z.; Pleash, J.; Helms, A.; Greaves, K.

2025-05-21 health systems and quality improvement
10.1101/2025.05.21.25328063 medRxiv
Show abstract

BackgroundLate discharges in cardiology inpatient units contribute to patient flow inefficiencies, emergency department overcrowding, and reduced hospital capacity. We define late discharges as those that occur after 12:00pm. MethodsWe conducted a quality improvement initiative using the Plan-Do-Check-Act framework over the course of one year. We evaluated a baseline mean discharge time in patients over the course of one month and identified causes of delays. Then, through iterative qualitative analysis we generated and implemented new departmental policies aimed at improving discharge times. We then re-assessed discharge outcomes at one, and six months post intervention. ResultsFollowing intervention, we saw a reduction in median discharge time of 1h 27m (95% CI 0:11-2:14, p<0.01) after one month compared to baseline. This was sustained at after six months with a 1h 20m reduction (95% CI 0:23-2:10, p<0.01), with no significant difference between the one and six month periods. Participants were also more likely to have a timely discharge before 12:00 at one (RR=3.03, 95% CI 1.84 - 4.99. p<0.01) and six (RR=2.39, 95% CI 1.43 - 4.01. p<0.01) months, compared to baseline. ConclusionTargeted interventions significantly improved discharge efficiency in a cardiology inpatient unit, enhancing patient flow and operational performance. This framework can inform similar initiatives in other settings.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.